I wasn’t expecting this much discussion. I originally had this at $4, but realized pretty early on it cannot cost $4. An easy, quick comparison is
Ironmonger. You get a fixed +1 Card, +1 Action, and half Survivors, along with one of +1 Card, +1 Action, and +$1, which are the best bonuses to spam.
But in case you want to go into all the details, the following is a list of all the raw possibilities along with their probability of occurrence and estimated value.
You might notice that certain individual effects are valued higher than you would think; we’ll get into more of that later.+3 Cards (0.612%) Smithy, but better (~$5.5)*
+2 Cards +1 Action (2.296%) Lab ($5)
+2 Cards +1 Buy (2.296%) Silk Merchant, but better?!? ($5.3)*
+2 Cards +$1 (2.296%) Fortune Hunter, but better (~5.6)*
+2 Cards Survivors (2.296%) Oracle, but better (~$5.4)*
+1 Card +2 Actions (2.296%) Village ($3)
+1 Card +1 Action +1 Buy (5.102%) worse than Market Square (~$2. 8 )
+1 Card +1 Action +$1 (5.102%) peddler (~$4.5)
+1 Card +1 Action Survivors (5.102%) most comparable to Spy (~$3.6)
+1 Card +2 Buys (2.296%) multiple +Buy bad (~$3.3)*
+1 Card +1 Buy +$1 (5.102%) not a lot of reference (~$3. 8 ) *
+1 Card +1 Buy Survivors (5.102%) not a lot of reference (~$3.4)*
+1 Card +$2 (2.296%) could maybe justify this as a terminal Silver+ (~$3.9)*
+1 Card +$1 Survivors (5.102%) not a lot of reference (~$3.6)*
+1 Card Survivors x2 (2.296%) not a lot of reference (~$3.2)*
+3 Actions (0.612%) strictly worse than Port, comparable to Snowy Village and Crossroads (~$3)
+2 Actions +1 Buy (2.296%) not a lot of reference (~$2.5)
+2 Actions +$1 (2.296%) worse than Merchant Camp (~$2. 8 )
+2 Actions Survivors (2.296%) not a lot of reference (~$2.3)
+1 Action +2 Buys (2.296%) multiple +Buy bad (~$1.5)
+1 Action +1 Buy +$1 (5.102%) worse than Candlestick Maker (~$1.9)
+1 Action +1 Buy Survivors (5.102%) bad, but not a lot of reference (~$1.7)
+1 Action +$2 (2.296%) Silver-ish (~$3)
+1 Action +$1 Survivors (5.102%) bad, but not a lot of reference (~$1. 8 )
+1 Action Survivors x2 (2.296%) bad, but not a lot of reference (~$1.5)
+3 Buys (0.612%) multiple +Buy bad (~$1.2)
+2 Buys +$1 (2.296%) worse than Squire (~$1.9)
+2 Buys Survivors (2.296%) multiple +Buy bad (~$1.5)
+1 Buy +$2 (2.296%) Woodcutter ($3)
+1 Buy +$1 Survivors (5.102%) bad, but not a lot of reference (~$1. 8 )
+1 Buy Survivors x2 (2.296%) bad, but not a lot of reference (~$1.3)
+$3 (0.612%) probably slightly better than Fortune Hunter (~$4.2)
+$2 Survivors (2.296%) Worse than Navigator, which was already bad (~$2.7)
+$1 Survivors x2 (2.296%) (~$1.4)
Survivors x3 (0.612%) Bad (~$1.1)
Remember not to make the card a 3$ if you think it's stronger than the average 4$.
Like I said, it’s easy to think of the Labs and Smithies and Peddlers and forget all the other junk that’s possible.
If you tally the weighted value it’s about $3. Weighted value of everything above the line (with +Cards) is about $4, and weighted value of everything below the line (without +Cards) is about $2.
A lot like Knights, the value of Redoubt fluctuates based on what the top card of Ruins is at the start of the game. Ruined Library and Ruined Mine provide the most value. Ruined Village is good in the absence of a Village, and Ruined Market is great in the absence of +Buy. Survivors is only good if the kingdom has no trashing and little draw. So you can use the knowledge of the top card to adjust the probabilities and be more informed about the value of the card.
For example, say it’s a 2-player game without Ruined Library on top. The chance of NOT having Ruined Library in the next 2 cards is about 80%, so if you want your Redoubt drawing cards (the most valuable effect), then you have to get decently lucky. If you really want to dig into the Ruins pile for that Ruined Library, you need to gain 3 Redoubts and collide them in the same hand (difficult in itself). You’re most likely to get 1 Ruined Library (43%), with the next likely outcome being no Ruined Libraries (34%). That leaves a (23%) of hitting 2 Ruined Libraries and those spicy $5+ effects. Not very good odds.
Any combo using Ruined Library and other terminals is actually better than its face value because you can play some Ruined Libraries to potentially draw more villages before playing your last terminal card. I already adjusted the card values for that above. (*)
Everything below the line doesn’t have +Cards, so it tends to be worth a lot less. You may notice I rate the bad terminal stop-card combinations a little higher than they deserve face value. That’s because you always have the option to not play the Ruins and keep the innate +3 Actions, which I’m valuing at about $3. Then you could play other terminal draw for the chance at drawing more villages while risking junking yourself. Or you could just play terminal payload and take the junk.
The notion that the opponent might gain and trash Ruins seems dubious, self-junking is after all a heavy price to pay.
Removing Ruined Library as an option lowers Redoubt’s value to about $2, so it’s worth it if your opponent has a ton of them and is using the draw a lot. It’s kind of like when you have 1 Poacher and the opponent has 8; you just gain the last Poacher to turn them into Oasis. But here you don’t need to gain extra Redoubts or Ruins. Just play one Redoubt (you probably got one early on), get the Ruined Library that’s 3 cards down, and that’s it.
And at worst you still got a Double-Necro whereas he had to self-junk & trash for little benefit.
To get the double Necropolis effect for ~$3, the opponent has to trash or keep in their deck the Survivors and other terminal Ruins, so that’s not free either.
On a serious note, I'd consider returning the card to the bottom of the pile
Returning the ruins to the bottom of the pile is something I hadn’t considered, and could probably work. My worry is that it makes it too unpredictable early and too uncontrollable late. Remember the average effect is actually $3, and you have to work to make it worth more than that. Also
Black Market is supposed to put cards on the bottom of the pile, but dominion.games doesn’t do that because then you have to track the pile (I think TGG lets you see the card order, but I’ve never been in a game that when through the whole pile to find out for sure). But the Ruins pile is smaller unless you have a ton of players, so tracking is less of an issue here. I’d probably want to playtest it both ways to see how it works.